1. Dynamics of A Three-Variable Nonlinear Model of Vasomotion: Comparison of Theory and Experiment 
D. Parthimos, R.E. Haddock, C.E. Hill, T.M. Griffith 
Biophysical Journal 
Volume 93, Issue 5, Pages (September 2007)
DOI: /biophysj
Copyright © 2007 The Biophysical Society Terms and Conditions

2. Figure 1
Nonscaled steady-state open probability curves. (A) The RyR curves are fourth-order Hill sigmoidals whose sensitivity to Ca2+ is governed by the coefficient xr, which defines the concentration producing half-maximal activation of the channel. (B) Sample InsP3R curves plotted for three values of the half-point of the activation sigmoidal (xi) and three values for the half-point for the inactivation sigmoidal (hi). The Hill coefficient for both sigmoidals was selected as 4. Parameters were selected to encompass the range of positions of the RyR and InsP3R curves reported in the literature (see text). In general, the maximum open-state probability of the channels may be <1, but this is accounted for by the scaling coefficients Cr and Ci.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

3. Figure 2
Four approaches simulating the reported actions of ryanodine by changing coefficients in Eqs. 1 and 2. (A) Increases in Cr from 1250 to (B) Decreases in the Ca2+ sensitivity coefficient xc from 0.9 to 0.6. (C) Increases in L from to (D) Reductions in Cr from 1250 to 500. (E–G) Experimentally, ryanodine increased oscillatory frequency in association with a reduction in the amplitude of oscillations in diameter, membrane potential, global wall [Ca2+], and [Ca2+] in individual smooth muscle cells (SMC), but had little effect on their average values.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

4. Figure 3
Nullcline analysis of the putative actions of ryanodine. (A–C) Pairs of nullclines and corresponding limit cycles for changes in parameters Cr, xc, and L. The intersection of the x (dashed line) and y (solid line) nullclines determines the equilibrium points indicated by an open circle when unstable (i.e., oscillatory) and a solid circle when stable (nonoscillatory). In each case, progressive decreases in the coefficient under investigation ultimately suppressed oscillatory behavior. (D) Effects of these interventions on oscillation frequency and amplitude.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

5. Figure 4
Effects of changes in the RyR Hill coefficients pr and mr. (A) Graded reductions in the activation coefficient pr from 4 to 2 reduced the size of the limit cycle until oscillatory behavior ceased. (B) Reductions in the Ca2+ release coefficient mr progressively increased the size of the limit cycle but did not stop oscillations. Dashed and solid lines denote the x and y nullclines; open and solid circles denote unstable and stable equilibrium points.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

6. Figure 5
Nullcline analysis of the dynamics of InsP3R-mediated CICR. (A–C) Pairs of nullclines and corresponding limit cycles for different values of the open probability coefficient Ci, activation sensitivity xi and inactivation half-point hi. Dashed and continuous lines denote the x and y nullclines; open and solid circles denote unstable and stable equilibrium points.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

7. Figure 6
Nullcline analysis for the activating and inactivating InsP3R Hill coefficients. Plots are shown for (A) the activation coefficient pi,, (B) the store release coefficient mi, and (C) the inactivation coefficient qi. Dashed and solid lines denote the x and y nullclines; open and solid circles denote unstable and stable equilibrium points.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

8. Figure 7
Analysis of combined RyR- and InsP3R-mediated CICR. (A) Progressive activation of Ca2+ release by ryanodine (simulated as an increase in Cr) in the presence of fixed InsP3-mediated CICR (constant Ci) can initiate or suppress oscillatory activity. (B) Similar analysis for progressive increases in coefficient Ci for fixed Cr. (C and D) Two-dimensional projections of the full three-dimensional attractor of the dynamics. Increases in coefficients Cr and Ci translate the attractors in the direction of the arrows. The inset in B illustrates the relationship between the equilibrium values [Ca2+]i (x*) and membrane potential (z*) as a solution of Eq. 4. Dashed and solid lines denote the x and y nullclines; open and solid circles denote unstable and stable equilibrium points. Nullclines and limit cycles were obtained with z held at −40mV; attractors were generated from trajectories where z was a free variable.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

9. Figure 8
(A) Reductions in term A of Eq. 1 were used to simulate the action of U73122 (from A=2.3 to A=0.0). (B) Experimentally, 10μM U73122 caused membrane hyperpolarization, dilatation and loss of rhythmic activity. (C and D) 10μM U73122 also inhibited oscillations in global wall [Ca2+] and cytosolic [Ca2+]i in individual smooth muscle cells.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

10. Figure 9
(A and B) Reductions in the coefficient GCa of the VOCC term in Eqs. 1 and 3 from 12 to 0 were used to simulate the action of nifedipine. When coefficient A was relatively large (2.3 (A)) blockade of VOCCs did not suppress oscillatory activity, whereas for small values of A (0.7 (B)) oscillations were abolished. (C–E) Experimentally, 1μM nifedipine effectively abolished vasomotion with oscillations in wall [Ca2+] and in individual smooth muscle cells being markedly reduced in amplitude.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

11. Figure 10
Bifurcation analysis of [Ca2+]i oscillations constructed by varying the parameters GCa (VOCCs) and A (NSCCs). (A) GCa was varied from 30.0 to 0.0, whereas A was varied from 0.1 to 3.3. Decreases in GCa demonstrate the occurrence of a Hopf bifurcation followed by a reverse Hopf bifurcation when GCa is further decreased. The reverse Hopf bifurcation occurs outside the physiological range (i.e., GCa<0) for large A values, thus indicating that oscillatory activity may either be abolished (low A) or sustained (high A) after blockade of VOCCs. (B) Corresponding time series illustrating the oscillatory and nonoscillatory domains in GCa-A parameter space.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

12. Figure 11
Nullcline analysis illustrating the stability characteristics of the system under the variation of (A) parameters Cr and GCa (with A=0.5), and (B) parameters Cr and A (with GCa=12). Dashed and solid lines denote the x and y nullclines; open and solid circles denote unstable and stable equilibrium points.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

13. Figure 12
A reduction in the coefficient GK of Eq. 3 was used to simulate the action of charybdotoxin. (A) Oscillatory activity was almost abolished when GK was reduced from 43 to 28, in association with marked membrane depolarization and increased [Ca2+]i. (B) Experimentally, charybdotoxin (60 nM) markedly reduced the amplitude of contractile and electrical activity and caused membrane depolarization. The residual activity exhibited an increase in frequency. (C and D) Oscillations in wall Ca2+ and in individual smooth muscle cells were similarly increased in frequency with a rise in mean [Ca2+].
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

14. Figure 13
(A) Action of sodium isethionate simulated by introducing a stepwise change in the Cl− channel reversal potential zCl by a ΔzCl of 17mV, which resulted in significant membrane depolarization, a slight rise in average [Ca2+]I, and suppression of oscillatory behavior. (B–D) Experimentally, chloride substitution with 120mM sodium isethionate caused depolarization and abolished rhythmic contractile activity and Ca2+ oscillations and also caused a small increase in mean [Ca2+]i. (E) Simulation of the action of chloride channel blocker niflumic acid by reducing coefficient GCl from 65 to 0 (with A=1.0) resulted in hyperpolarization and a reduction in [Ca2+]i. (F–H) Experimentally, niflumic acid resulted in hyperpolarization and a reduction in [Ca2+]i with an associated loss of rhythmic activity.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

15. Figure 14
(A) Over the physiological range, individual components contributing to membrane potential include K+ efflux, which promotes hyperpolarization, whereas Cl− efflux and Ca2+ influx through VOCCs both promote depolarization. The role of Na+/Ca2+ exchange is variable because of the Ca2+ dependence of the reversal potential of this exchange mechanism. A hypothetical composite dz/dt plot (see Eq. 3) shows that three possible steady states occur at points of intersection with the horizontal axis (P1 and P3, stable; P2, unstable). (B) Family of dz/dt plots encompassing the physiological range, with each plot corresponding to a distinct value of [Ca2+]i at intervals of 0.05μM, as indicated. The plots provide insight into the hyperpolarizing action of U73122 as this agent decreases [Ca2+]i. (C) The effects of nifedipine were modeled as a reduction of GCa to generate two families of dz/dt plots, each covering the range [Ca2+]i=0.1–0.9μM. Theoretical effects of nifedipine on membrane potential are indicated by arrows at the same value of [Ca2+]i and show that hyperpolarization is most marked in depolarized vessels when [Ca2+]i is high. (D) Analogously, the depolarizing effects of charybdotoxin were modeled by reducing GK to generate two families of plots encompassing different values of [Ca2+]i. The corresponding depolarizations are indicated by arrows. (E and F) A similar approach was used to understand the role of Cl− channels after substitution by isethionate or administration of niflumic acid. These simulations respectively demonstrate depolarization and hyperpolarization.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

16. Figure 15
(A) U73122, simulated as stepwise reductions in term A, promotes membrane hyperpolarization and suppression of oscillatory activity. (B) Ryanodine, simulated as stepwise increases in coefficient C, is associated with a gradual reduction in oscillatory amplitude as [Ca2+]SR is reduced and stores are gradually depleted. There is minimal effect on average [Ca2+]i and membrane potential. (C) Nifedipine, modeled as stepwise decreases in coefficient GCa, causes a reduction in both [Ca2+]i and membrane potential. (D) Charybdotoxin, simulated as decreases in coefficient GK of Eq. 3, induces marked depolarization and an increase in [Ca2+]i. (E) Chloride substitution by isethionate, modeled as changes in zCl, produced marked membrane depolarization, associated with a small increase in mean [Ca2+]i and ultimate suppression of oscillatory activity. (F) Chloride channel blocking by niflumic acid produced hyperpolarization and reduced [Ca2+]i. Arrows indicate the direction in which the attractor translates during intervention.
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions

17. Figure 16
(A) Simulated action of charybdotoxin for two levels of Ca2+-dependence of KCa channels. Four stepwise changes in GK are included in each time series plot (GK=43, 37, 29, and 25). (B) Simulated time series for the action of isethionate for two levels of Ca2+ dependence of Cl− channels. Four stepwise changes in zCl are included in each time series plot (ΔzCl=0, 10, 18, and 20mV). (C) Simulated action of niflumic acid under the same conditions as in B. Stepwise changes in GCl are included in each time series (GCl=65, 35, 10, and 0, for A=0.9).
Biophysical Journal  , DOI: ( /biophysj )
Copyright © 2007 The Biophysical Society Terms and Conditions